Designing a Neural Network Control System

In summary, a user is seeking advice on designing a neural network based control system for managing power and temperature in a set of 120 computer servers. The desired temperature is 35 degrees Celsius and the goal is to reduce power in areas with the highest temperature difference until reaching a minimum heated state. However, it is recommended to use a simple PI controller instead of neural networks for this task. Resources for creating and tuning a PI controller can be found through a Google search.
  • #1
date.chinmay
10
0
Hello all
I'm trying to design a Neural network based control system which will do the following
i have 120 computer servers with
a. power
b. temperature of server inlet
c. temperature of server outlet

Now i have readings for all 3 operating from 5kw to 35 kw in steps of 5 kw
My server desired temp is 35 deg celcius.

what i want to do is my control sys should check the temperatures.. and where it finds highest delta.. it should reduce power by unit.

and go on doing this till it reaches minimum heated state.

Any ideas?
 
Last edited by a moderator:
Engineering news on Phys.org
  • #2
Neural networks is not the right way to do that.

A simple PI controller is the best. A Google search can show you how to make a PI controller and how to tune it.
 

1. What is a neural network control system?

A neural network control system is a type of artificial intelligence that uses a network of interconnected nodes to process and interpret data, make decisions, and control a system or process. It is inspired by the structure and function of the human brain.

2. How does a neural network control system work?

A neural network control system works by receiving input data, processing it through multiple layers of interconnected nodes, and producing an output that controls a system or process. The network learns and improves its performance through training on a large dataset.

3. What are the advantages of using a neural network control system?

Some advantages of using a neural network control system include its ability to handle complex and nonlinear relationships between inputs and outputs, its ability to learn and adapt to new data, and its potential for high accuracy and efficiency.

4. What are some key considerations when designing a neural network control system?

When designing a neural network control system, it is important to consider factors such as the type and amount of data available, the structure and size of the network, the choice of activation functions and training algorithms, and the overall goal and performance requirements of the system.

5. Can a neural network control system be used in any application?

While neural network control systems have a wide range of applications, they may not be suitable for all types of systems or processes. It is important to carefully evaluate the specific requirements and limitations of a system before deciding if a neural network control system is the best solution.

Similar threads

Neural Network based controller
    • Mechanical Engineering
Replies
2
Views
1K
Designing an Actual Vapor-Compression Refrigeration Cycle
    • Mechanical Engineering
Replies
8
Views
2K
How to control turbine in Organic Rankine Cycle system
    • Mechanical Engineering
Replies
3
Views
1K
Seawater sample temperature control – heat exchanger design
    • Mechanical Engineering
Replies
12
Views
2K
Neural Network based controller
    • Electrical Engineering
Replies
4
Views
1K
Automotive Calculation of energy storage system for PEV
    • Mechanical Engineering
Replies
1
Views
1K
System design for charging battery from turbine alternator
    • Electrical Engineering
Replies
8
Views
983
MHB Convexivity issues in a new type of network - NLP
    • Programming and Computer Science
Replies
1
Views
2K
Types of signals in process control
    • General Engineering
Replies
2
Views
4K
Discrete Control System with Time Delay
    • General Engineering
Replies
1
Views
4K

Hot Threads

Recent Insights

Back
Top